New paper – Transparency or Efficiency? The Prevalence of Arbitration Clauses in Croatian Public Procurement Draft Contracts
New paper out! I spent weeks scraping 17,695 public procurement notices from Croatia’s electronic procurement platform (EOJN) and going through 6,409 draft contracts to answer a question nobody has empirically tackled in Croatia before: how often do contracting authorities actually choose private arbitration over regular courts for resolving public procurement contract disputes? The short answer is 5.32% — 341 draft contracts with a combined estimated value of over 639 million EUR. That might not sound like a lot percentage-wise, but consider what it means in practice: hundreds of contracts funded by public money are being channelled into confidential proceedings where there’s no publicly available database of decisions, no obligation to disclose anything, and absolutely no public law safeguards tailored to the fact that one of the parties is a public body spending taxpayer funds.
What makes Croatia such an interesting case is that it’s a textbook droit administratif system — procurement procedures are administrative procedures, the appeal body (DKOM) is an administrative authority, judicial review goes to the High Administrative Court — but the moment you get to the contract itself, everything flips to civil law. The Supreme Court said in 2022 that public procurement contracts are administrative contracts, and yet not a single one of the 6,409 draft contracts I analysed references administrative courts as the competent forum. Not one. Every single non-arbitration contract points to civil or commercial courts.
The data threw up some patterns I genuinely didn’t expect. The Ministry of Interior alone accounts for 138 of the 341 arbitration clauses — that’s 40.5% of the entire dataset and around 165 million EUR in contract value, spread across goods, services, and works. One ministry, nearly half the dataset. And 97.1% of all contracts with arbitration clauses designate the exact same venue: the Permanent Arbitration Court at the Croatian Chamber of Economy (PAC-CCE). The only real alternative is the ICC, but those 8 contracts belong exclusively to public companies and utilities running major infrastructure and energy projects, with an average value of 17.5 million EUR versus 1.5 million for PAC-CCE contracts. So what you’re looking at is an informal two-tier system — domestic arbitration for everyone, international arbitration for a handful of commercially sophisticated state-owned entities doing big-ticket deals.
The cost-proportionality question is where it gets really uncomfortable. Over 68% of these contracts are valued below 500,000 EUR and 16% are under 50,000 EUR. Run those numbers through the PAC-CCE fee calculator and you’re looking at arbitration costs of up to 15.67% of the dispute value for a 50,000 EUR contract with a three-member tribunal. That’s not efficiency — that’s what the literature calls “passive waste,” where standardised templates get copy-pasted across contracts without anyone stopping to ask whether arbitration actually makes financial sense for a procurement worth less than a mid-range apartment. France, Portugal, and Italy have all developed specific rules to deal with this: mandatory publication of arbitral awards, ministerial approval requirements, the right of successful tenderers to refuse arbitration clauses, appeal rights to supreme administrative courts. Croatia has none of it. The paper argues it’s time to change that — and that reforming arbitration might actually be the most practical route to finally settling the long-running doctrinal debate about whether public procurement contracts in Croatia are administrative or civil law contracts.
Paper is up on SSRN: